SeeDance 2.0 Fast (BudgetPixel)
SeeDance 2.0 Fast — the same Seedance 2.0, at half the resolution ceiling
- audio
- end_image
- image
- reference_images
- video
- video
SeeDance 2.0 Fast is what it says on the tin: the same four-mode SeeDance 2.0 (text-to-video, image-to-video with start/end frames, reference-to-video, video-to-video editing) at lower cost and latency - 160 credits per second against the standard 220. The trade is that it tops out at 720p; there's no 1080p or 4K in the resolution dropdown. For drafts, iterating on prompts, or any output that's going to be compressed down for social anyway, that's a fine bargain.
What's different from SeeDance 2.0
Nothing structurally - same mode logic, same mutually-exclusive input rules, same video edit socket with the same input-duration-at-half-rate billing on top of the output. The only real differences are price, latency, and the resolution ceiling. If you've already read the SeeDance 2.0 article, you know this node. The decision between the two is purely economic: 2.0 Fast for anything that doesn't need to come out at 1080p, 2.0 for the high-res hero clips.
The inputs that matter
- prompt - required; the edit instruction when
videois connected. - image / end_image - start and optional end frame for image-to-video.
- reference_images - up to 9 for reference-to-video; can't combine with
image/end_image. - video - the clip (2–15s, ≤100MB) for video-edit mode.
- length_seconds - 4–15.
- resolution - 480p or 720p. That's the whole list.
- generate_audio - off by default.
Output is video (VIDEO), core type - needs 2025+ ComfyUI.
How it works
The pack's standard hosted flow: media serialized and uploaded, a job created, polled until done, MP4 downloaded into the temp folder and wrapped as a VIDEO value. Video jobs take minutes; the "fast" is relative to the standard model's latency, not a local render. Interrupting the graph stops the wait but the job keeps running and billing server-side.
Install and key setup
ComfyUI Manager → BudgetPixel, or:
cd ComfyUI/custom_nodes
git clone https://github.com/budgetpixel-ai/comfyui-budgetpixel.git
pip install -r comfyui-budgetpixel/requirements.txt
Only requests. API key from a paid plan via BUDGETPIXEL_API_KEY or budgetpixel.json; never in workflow JSON.
Where people get burned
People pick Fast and then wonder why there's no 1080p option - read the resolution list before you commit, because the standard 2.0 node is the only route to high-res Seedance. The video-edit double-billing still applies (input duration at half rate plus output), so cost-estimate before an edit job. And the pack rules hold: force_rerun to defeat caching, paid plan required for the API.
Inputs (11)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | Text description of the video. | |
| aspect_ratioopt | COMBO | 16:9 | Aspect ratio. Ignored when an input image or video already sets the frame. |
| audioopt | AUDIO | Optional reference audio (MP3/WAV/OGG) for the generated video. Provide a public audio URL or an uploaded-file URL from POST /v1/uploads (passed by URL, not inlined). Must accompany an image (`image` or `reference_images`) or a `video` input — audio can't be the only input. | |
| end_imageopt | IMAGE | Optional end frame, used together with `image` (the start frame) to interpolate the video between the two frames. Same input forms as `image`. | |
| generate_audioopt | BOOLEAN | false | Generate audio with the video. |
| imageopt | IMAGE | Optional start frame for image-to-video. Provide a public image URL, a data URI, raw base64, or an uploaded-file URL from POST /v1/uploads. Omit for text-to-video. | |
| length_secondsopt | INT | 54–15 | Video length in seconds (4–15). |
| reference_imagesopt | IMAGE | Optional reference images that guide the generated video. Each item is a public image URL, a data URI, raw base64, or an uploaded-file URL from POST /v1/uploads. Can't be combined with `image`/`end_image` (start/end frames). Up to 9 on their own (reference-images-to-video); may also accompany a `video` to guide a video-edit, in which case up to 6 are used. | |
| resolutionopt | COMBO | 720p | Output resolution. Pricing varies by resolution — see `resolution_pricing` in GET /v1/models. |
| videoopt | VIDEO | Optional input video for video-to-video editing (2-15s, ≤100MB). Provide a public video URL or an uploaded-file URL from POST /v1/uploads (videos are passed by URL, not inlined). When set, the request runs in video-edit mode; you may also pass `reference_images` (up to 6) and/or an `audio` track to guide the edit. Not combinable with `image`/`end_image`. Billed on the INPUT duration at half the output per-second rate in addition to the output — estimate via POST /v1/cost. | |
| force_rerunopt | BOOLEAN | false | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| video | VIDEO | — |